With a little help from a friend: A Shower Calendar to save water

Abstract

This design case presents and discusses the Shower

Calendar, a "persuasive" concept for reducing the con-

sumption of water for showering. It starts from a dis-

cussion of different types of feedback employed by

earlier design cases. Based on this, we designed the

Calendar concept as an ambient, persistent and indivi-

dualized feedback. A field study with two families (6

individuals) revealed that the Calendar fosters goal

setting, comparison, competition, and communication.

In addition, quantitative data showed one family to

have been more successful in translating the Calendar's

offer into actual behavior change, i.e., saving water.

This highlights that change is not achieved by the

product itself (as in automation or regulation), but by

the people involved.

Keywords

User experience, experience design, persuasive tech-

nology, behavior and attitude change, sustainability

ACM Classification Keywords

H5.m. Information interfaces and presentation (e.g.,

HCI): Miscellaneous.

General Terms

Human Factors, Design

Copyright is held by the author/owner(s).

CHI 2011, May 7–12, 2011, Vancouver, BC, Canada.

ACM 978-1-4503-0268-5/11/05.

Matthias Laschke

Folkwang University of the Arts

45141 Essen, Germany

matthias.laschke@folkwang-uni.de

Marc Hassenzahl

Folkwang University of the Arts

45141 Essen, Germany

marc.hassenzahl@folkwang-uni.de

Sarah Diefenbach

Folkwang University of the Arts

45141 Essen, Germany

sarah.diefenbach@folkwang-uni.de

Marius Tippkämper

Folkwang University of the Arts

45141 Essen, Germany

tippkaemper@googlemail.com

2

Introduction

Freshwater is – like any other natural resource – pre-

cious. Not surprisingly, there is a growing interest in

products and systems, which aim at reducing our con-

sumption of water and other resources. One domain for

this endeavor is the domestic. Labeled as sustainability

and eco-minded design, nowadays consumers are con-

fronted with water saving dishwashers and washing

machines, automated lighting controls or even more

elaborated "smart homes" and "grids." Many of those

products seek to reduce energy consumption through

increased efficiency or automation. They typically do

not involve their users actively beyond the buying deci-

sion. But even the buying decision is increasingly re-

placed by regulation. After the ban of regular light

bulbs, the European Union now plans for a regulation of

shower heads, which "waste" water.

The ban of particular "bad" products as well as the

appreciation of "smart homes" seems to – at least im-

plicitly – assume that people are not willing to actively

participate in resource conservation. This is interesting,

given the fact that research since the 80ties of the last

century shows that direct and indirect feedback of

energy consumption has a significant effect in terms of

habit change (see [5] for an overview). In addition,

there are rising doubts whether, for example, energy

efficiency of lighting really leads to a decrease in ener-

gy consumption. The opposite may be true due to an

increase in general light consumption enabled by more

efficiency (see for example

http://thebreakthrough.org/blog/2010/09/why_energy

_efficiency_does_not.shtml). Consequently, there

seems to be a demand for well-designed systems,

which keep their users "in the loop" instead of deferring

saving to materials, automation, or regulation.

Interactive products are an ideal medium to foster such

behavior and attitude change [7]. They offer a wide

array of customizable real-time feedback and allow for

immediate responses to this feedback. Accordingly,

"persuasive technologies" became an important field of

research within Human-Computer Interaction. Howev-

er, available design cases reveal a wide variety of po-

tential solutions, all seemingly based on the similar

broad principles, but very different in detail. In other

words, while the broad principles remain the same,

their materializations as products substantially vary.

Following the idea of persuasive technologies as "mate-

rialized arguments" in a dialog between user and de-

signer [17], these variations will surely impact the "sto-

ry" told by the product, its perception and ultimately its

meaning and its power to change attitude and beha-

vior. Collections of principles, such as feedback, goal

setting, comparison [8], are very helpful to generally

understand how behavior can be changed. But they

neither prescribe nor discuss in detail, how these prin-

ciples should be met by a concrete product. This makes

the exploration of design cases as concrete materializa-

tions of underlying persuasive (i.e., psychological) prin-

ciples worthwhile.

This design case presents and discusses the Shower

Calendar, a concept for reducing the consumption of

water for showering – an interesting application domain

for persuasive technologies. Bathing and showering

makes up a large part of domestic water consumption,

for example, 36% of the total water consumption of a

typical German household [18]. Moreover, water con-

sumption provides direct opportunities for saving. A

fridge, for example, can't be used less frequently. It is

either on or off. There is some potential for saving,

such as not putting hot meals into the fridge or buying

3

a new one. But the actual potential for improved beha-

vior is rather small. This is different for water consump-

tion, which consists of frequent, small acts like shower-

ing, dish-washing, and so forth.

We start this case with a detailed discussion of different

types of feedback and their potential drawbacks. The

impatient reader may directly turn to the subsequent

description of the Shower Calendar concept and our

quantitative and qualitative field study of a functional

prototype.

Feedback

From judgmental to ambient

A basic feature of persuasive technologies (and a cor-

ner stone of behavioral change in general) is appropri-

ate feedback about resource consumption (see [5]) to

create "energy awareness" and to prompt behavioral

change. However, feedback can take many different

forms. For example, UpStream [12] uses a "traffic

light"-metaphor. A green light is displayed as long as

the consumption is below the average of everything

previously measured. When the user is about to exceed

the average, the light changes to yellow. It becomes

red, if the water has been running for longer than one

standard deviation above average. Finally, it even

flashes (thereby breaking the "traffic light" metaphor).

This type of feedback is rather judgmental. It bluntly

categorizes some behavior as "good" (green) and other

behavior as "bad" (red), but ignores the possibility that

there might be a legitimate reason for using more wa-

ter. This runs the risk of being too pushy and even

unfair, which might rather lead to reactance [2] than

change.

Another kind of feedback widely used is positive rein-

forcement – rewards, points or trophies. Waterbot [1],

for example, is using positive auditory messages and

chimes when closing the tap. It monitors and judges

behavior, similar to a punishing and praising dog train-

er. Besides the danger of being unnerving, this type of

feedback rather externalizes the reasons for behavior.

One can submit oneself to Waterbot – she will guide me

and take on the responsibility. However, to be persis-

tent, behavior and reasons for this behavior have to be

internalized. I should change because I want to, not

because Waterbot says so.

A more subtle road is ambient feedback. The Power

Aware Cord, for example, visualizes the current uptake

of electricity of a connected appliance through glowing

pulses, flow, and intensity of light [9]. This is not as

judgmental as a traffic light and also may be subtle

enough to be ignored, if increased consumption ap-

pears legitimate. Show-me [11] visualizes the water

consumption by a row of LEDs on a bar in the shower

cabin (Figure 1).

Figure 1. LED-bar prototype of Show-me (left) and a further

integrated design solution (right) [11].

4

For every five liters consumed, an additional LED is

switched on. As a consequence, the shower cabin be-

comes a brighter place with increasing water consump-

tion. While certainly being ambient, this form of feed-

back seems to beautify consumption rather than to

criticize it. Ambient displays must tell an appropriate

story, namely one of loss over time rather than gain.

Thus, the exact opposite – a decrease in light – may

have been much more appropriate in a dialog about the

responsible use of a limited resource.

Consumption patterns and a record of change

Feedback in persuasive technologies is not only a mat-

ter of awareness of resource consumption at a given

moment, but also a matter of awareness of consump-

tion patterns over time. Show-me, for example, only

displays the amount of water used during one shower.

This direct feedback certainly calls for an immediate

response. However, the moment the user steps out of

the shower, the information is only available in his or

her memory. Upstream represents earlier behavior only

indirectly by the average the traffic light is centered on.

However, if the objective is to change behavior, con-

sumption patterns over time have to be made accessi-

ble to the users. This calls for a persistent feedback, a

record and display of each single act of consumption, of

each single shower taken. Only such a record enables

users to compare showers taken, and to become aware

of improvement or deterioration in water consumption.

The potential focus on the trend over time instead of

single acts also helps to accept variety in own behavior.

Even if one "fails" today (either deliberately, because I

earned it, or by inattention), there can be still an im-

provement overall. This refocus on the overall pattern

is a desirable one, given that many prolonged acts of

self-control fail because of an underlying "on-off"-

model, with one transgression letting the whole effort

appear worthless [16].

Individualization and accountability

Mundane acts, such as showering, are embedded in

social contexts, such as a family situation or a shared

flat. Other than Show-me [11], Waterbot [1] explicitly

uses this social context and addresses the principle of

social validation (e.g., [4]) by putting current consump-

tion in relation to the average household consumption.

While this anonymous form of social feedback may be

generally viewed as more "politically correct" than indi-

vidualized feedback, it appears less adequate in the

present context for at least three reasons. First, it ac-

tually reduces the accountability for one's own behavior

– the exact opposite of what the conscious use of re-

sources requires. Second, while it may reduce social

pressure, and, thus, may be less judgmental, it also

reduces the chance of communication among users.

However, this social comparison seems important for

personal goal setting. Third, failures, which remain

anonymous, are certainly more bearable. However, the

same lack of individualization may hinder the full relish

of a success, which includes feelings of acceptance and

appreciation by others.

All in all, we argue for a vague and ambient display,

which appropriately maps consumption to a visualiza-

tion, which avoids the beautification of waste, is persis-

tent (i.e., displays not only water used, but also con-

sumption patterns over time) and individualized. We

assume that this combination will encourage the accep-

tance of responsibility for one's own behavior without

feeling coerced by the product. It should also foster

communication among its users in a social context,

which is surely an important part of the persuasive

5

process. In the following, we discuss the Shower Ca-

lendar as a potential materialization of the envisioned

type of feedback.

The Shower Calendar

Description

The Shower Calendar consists of display installed in the

shower (Figure 2, see also

http://www.youtube.com/watch?v=AHezdVxbi7M for an

extended video figure). Users identify themselves by

pressing a button. A large dot of the user's personal

color appears at the current date in a calendar-like

matrix. The dot represents 60 liters. This amount is

roughly based on an estimated consumption of 20-40

liters per day and person in Germany [19]. We took the

average and added 30 liters to allow for more extreme

consumption situations. We further added a "free"

amount of four liters before the dot starts to get small-

er. It then shrinks until the user either stops the water

or the full amount of 60 liters has been spent. In the

latter case, the dot is very small, but still a visible

record of this shower taken (Figure 3, detail). The free

amount of four liters was introduced to avoid possible

frustration. Otherwise the visualization would never be

visible to a full extent, because showering without us-

ing any water is impossible. The amount itself is based

on the first authors experience while trying to achieve a

comfortable feeling of cleanliness with a minimum of

water. A series of ten showers resulted in an average of

four liters used.

Design rationale

The dot was used as a fuzzy, ambient representation of

the amount of water used per shower. It became

smaller with consumption to communicate the limited-

ness of resources.

Figure 2. Shower Calendar in the bathroom (concept and

video still from http://www.youtube.com/

watch?v=AHezdVxbi7M).

The dots are arranged in a calendar-like matrix, a well

known pattern for recording and monitoring over longer

periods. The different colors identify different people,

thereby providing an individualized feedback over time.

6

However, the color is arbitrary, no names or nick

names are displayed. It requires an explicit communi-

cation among users to determine individuals ("Who has

blue?"). In addition, the whole process requires an

initiation by pressing a button. On the one hand, this

allows for an explicit commitment, which will increase

awareness per se. On the other hand, it allows for

cheating (e.g., showering without initiation of the sys-

tem, "identity theft"), which clearly communicates the

active role of the users themselves and the role of the

calendar as a tool for self-help and not as a watchdog.

There is no point in cheating, if you want to change

yourself. All in all, the display is designed in a way that

beautifies reduced consumption of water and the par-

ticipation of all family members through an emerging,

colorful pattern (Figures 2 and 3). The cabin becomes

more colorful day by day, which is a more implicit and

more lasting reward than any immediate chime or later

trophy (as suggested by [1]).

Prototype

The functional prototype of the Shower Calendar con-

sisted of a flow meter, a micro-controller board, a

number pad, a computer, and a screen (instead of an

integrated screen in the shower cabin or a projection)

(see Figures 4 and 5). The flow meter was installed

between the faucet and the flexible tube of the shower-

head. For each liter the computer decreased the co-

lored dot on the screen. The number pad was con-

nected to the computer. All electronic equipment was

outside the shower. Each user got assigned a button at

the number pad. The screen displayed all different dots

at each day for one year. The screen was placed within

sight, so that users were able to keep track of the cur-

rent dot during showering.

Figure 3. Dots displayed in a calendar-like matrix. Each col-

umn represents a month, each line a day. The detail shows the

left-over of using more than 60 liters (small blue dot on the

right).

Study

Two families participated in the study, each consisting

of three individuals. Family 1: 50-year-old mother

(P1.1), 55-year-old father (P1.2), 20-year-old son

(P1.3). They lived in a single-family house in a small

German city (62.000 residents). Family 2: 50-year-old

mother (P2.1), 20-year-old son (P2.2), 14-year-old son

(P2.3). They also lived in a single-family house in the

same city as Family 1. Both families were of a German

middle class, non-academic background. The son of

Family 1 was in an educational program. The younger

7

son of Family 2 attended public school. The older son

was about to begin his studies. Participants were asked

to use the Shower Calendar in their daily life for a pe-

riod of one month (31 days). The prototype of the

Shower Calendar was placed in the families' bathrooms.

The study started mid June 2010.

Figure 4. General setup of the functional prototype.

The Shower Calendar was briefly introduced to the

participants. This covered aspects, such as the personal

login before every shower, the differently colored dots,

and the amount of water a dot represents. However,

participants were not "educated" about water consump-

tion in general or strategies to save water. Naturally,

both families considered resource conserving as a posi-

tive and worthwhile aim. However, they neither in-

vested a lot of money nor time into it so far.

Figure 5. Specific setup of the functional prototype in Family

2's bathroom.

We focused our study on participants' experiences and

actual consumption. Accordingly, we led open, in-depth

interviews after the field trial, asking participants to

freely reflect on how the experience of showering had

become different (or not) with the Shower Calendar. To

stimulate participants, we prompted them – when ne-

cessary – with broad questions, such as "What were

your first thoughts and feelings when using the Shower

Calendar?", "Do you like the Calendar and if yes/no,

why?", "What do you think is the intention of the Ca-

lendar?", "Did the Calendar influence you and if so,

how?", "What do you think about products, which at-

tempt to change your attitudes or behavior?", "Do you

care about conserving resources and what do you do

for it?"

8

In addition, we logged the actual amount of water

spent per shower to get a complementary, quantitative

measure of actual behavior and its potential change.

Experiences

This section summarizes and exemplifies the topics

mentioned during the interviews in form of a narrative

explication.

Personal goal setting was repeatedly brought up:

"My first thought was how to save additional water. […]

Although I believe that my actual shower habits are not

that bad." (P1.1)

"Somehow it is about ambition: the ambition to shower

less." (P1.3)

However, it seemed important that these goals were

set by one self. This allowed for the adaption to situa-

tional needs, thereby reducing the feeling of being

coerced by the product. Broadly asked by the inter-

viewer, whether the Calendar influenced behavior and if

so, how, the mother of Family 1 remarked:

"I did not feel coerced to save water. For example, in

the last days it was so warm outside that I needed to

shower a little longer just for the refreshment." (P1.1)

"I did not feel patronized by the product, because I

could always choose for myself, how I wanted to be

influenced" (P1.1).

The Shower Calendar's design allowed for personal goal

setting by providing the size of the dot as feedback

about the amount of water consumed. However, this

unit is ambiguous, which accommodates various perso-

nally meaningful goals. This ambiguity makes the prod-

uct appear less judgmental. In addition, slightly ambi-

guous feedback allows for better ignoring the product's

feedback in situations in which using more water is felt

to be legitimate. The product must allow for these

"transgressions" to create the feeling of helping along

rather than patronizing or coercing.

An important aspect of goal setting is comparisons.

These can be comparisons between one's own

achievements over time and those of the others. In

Family 1, for example, the mother (P1.1) adopted the

role of the admonisher and the son (P1.3) was already

known for his water-wasting shower habits. Both were

fully aware of those roles.

The son reflected:

"At the beginning everybody grumbled at my small

dots. But when they became larger and larger, my

parents stopped. I was definitely influenced by them,

because we could see each other's dots." (P1.3)

And the mother said:

"We definitely monitored each other." (P1.1)

"My son proudly presented his dot to me: 'Look Mama!

Today my dot is even larger than yours'" (P1.1).

A similar theme of social comparison was apparent in

Family 2. The mother said:

"I had a look at the dots of the others [her children],

just to see whether my children take it seriously" (P2.1)

9

The possibility to compare also led to competition

among family members:

"I definitely monitored the dots of the others. […] There

is always a bit of a competition in a family." (P2.3)"

One mother stated:

"My husband always brags of his shower habits

'Soaping like a guy. Shower it off, done!' I wanted to

show him that I don't use much water either." (P1.1)

And further:

"At the beginning my husband and I laughed about [our

son's] small dots. […] Especially with teenagers and

kids it becomes competitive easily." (P1.1)

Also the father mentioned:

"Well, sometimes we laughed about the tiny dots of our

son." (P1.2)

But the son took on the challenge:

"It became similar to a game. In the beginning every-

body said 'You are the worst, you have the smallest

dots'. But in the end, I often had the largest dot."

(P1.3)

The Calendar also fostered communication. Shower

habits became a subject of family discussions:

"In the beginning, we now and then talked about the

shower during dinner." (P2.3)

The shower seemed to have a facilitating effect:

"After the installation of the shower, we talked much

more about water consumption. Normally, when I point

out to my children that they could consume less water,

it goes in one ear and out the other. […] But then, [my

son] told me about how fast his dots tend to disappear.

He was very astonished about it." (P2.1)

Interestingly, the shower became not only a subject

among the family members alone, but also among visi-

tors:

"Visitors asked me 'What is it for? Do you have Internet

in your bathroom?' I explained it to them." (P2.1)

Much of the shower's effect seems due to its ability to

foster communication. It becomes a communicative

tool, a tool for gaining insights and a way to make a

potential discussion less personal and accusatory. It

becomes a "material argument". And it is not taking

sides.

Experiences were accompanied by - at least intended -

behavioral changes. As one participant reflected:

"One changes behavior, because one knows that there

is the dot." (P2.1)

Because of this, people reconsidered and consciously

varied their actual shower behavior.

"I did not actually reduce the duration. To save water, I

just turned it off while soaping. […] I don't know,

whether I have done this before. But this time, I defi-

nitely did it on purpose." (P2.1)

10

Also the youngest participant pointed out:

"I did not reduce the duration, but I turned the water

off while soaping." (P2.2)

In sum, the participants did not feel coerced by the

Calendar, because it allowed for variability in behavior.

Through its fuzzy, ambient feedback participants were

encouraged to set personal goals, to compare them-

selves to others, and to monitor their progress. This

could even result in competition. In general, the Calen-

dar fostered and facilitated communication within the

families. Participants also directly reflected about beha-

vioral change.

Behavior

Figure 6 and 7 shows the Shower Calendar of Family 1

and 2 after the test period.

Figure 6. Visualization after test period (Family 1).

Figure 7. Visualization after test period (Family 2).

To complement the participants' insights with their

actual water consumption data, we calculated each

participant's average water consumption per shower

and the standard deviation (see Table 1).

The most interesting, however, is the change in the

amount of water used. To estimate this, we calculated

a linear regression for each participant, using time

(actually the sequence of measures) as predictor and

water consumption per shower in liter as criterion.

For four of the six participants the linear trend was

highly (P1.1, P1.2, P1.3) or at least marginally signifi-

cant (P2.1). For these four, the beta weight was nega-

tive, indicating a decrease of water consumption over

time. For the two sons in family 2 (P2.2, P2.3) there

was no significant linear trend. P2.3 did not change at

all. P2.2 even increased water consumption over time.

However, this participant only contributed 10 data

11

points (i.e., showers) over the 31 days, which leaves

the change insignificant.

Participant n M [l] SD [l] Change

( )

P1.1 (mother) 24 23.9 2.7 -.54**

P1.2 (father) 36 25.0 2.6 -.59***

P1.3 (son) 43 36.4 12.1 -.40**

P2.1 (mother) 21 28.0 2.2 -.40m

P2.2 (son) 10 25.6 3.1 +.52

P2.3 (son) 23 49.4 9.3 -.01

Notes: m) p<.10, **) p<.01, ***) p<.001

Table 1. Number of showers taken in the 31-day period (n),

mean amount of water consumed (M), standard deviation (SD)

and the change over time (β).

Beside the linear trend, we checked whether a satura-

tion curve would generally show a better fit to con-

sumption over time. This is, because water consump-

tion while taking a shower is not entirely avoidable

(only by not taking a shower at all). Actually, the

change of P1.1 and P2.1 (the mothers) was better

represented by a logarithmic than a linear function. The

shower calendar obviously encouraged them to save

water, but after they had reached their personal limit of

saving, water consumption remained at a constant

level.

Besides the differences in the shape of water consump-

tion over time, there were also significant differences

regarding the variability. Levene's test for Equality of

Variances showed that variability of water consumption

between the six participants was statistically significant,

F(5, 151) = 24.30, p < .001. Especially the son in fami-

ly 1 (P1.3) showed a highly variable showering beha-

vior.

The quantitative data highlights the complexity of fami-

lies as social entities. All parents showed a significant

(P1.1, P1.2) or marginally significant (P2.1) decrease of

water consumption over time, while the children either

conformed (Family 1) or not (Family 2). This can have

manifold reasons, but one is certainly the difference in

the assumed underlying wish to be transformed: the

children may simply not find it important to reduce

their water consumption (e.g., the 14 year old P2.2).

Some people are just more ready to change than oth-

ers [see 6]. Alternatively, the Shower Calendar's design

may be simply rendered inappropriate for people on

early stages of potential behavioral change [10]. This,

however, is then the consequence of the deliberate

design decision to address people already wishing to

change. While the present case is not able to answer

these open questions, it still reminds us of the com-

plexity of behavior and attitude change in a social set-

ting and questions approaches, which want to makes us

believe in the possibility to influence people in a simple,

passive and mechanistic way.

All in all, it becomes obvious that employing principles

like goal setting, comparison, competition, or commu-

nication cannot guarantee behavioral change. It is ra-

ther an offer to change, which has to be actively ac-

cepted. While some may view the latter as a serious

limitation, we understand a dialogue between the Ca-

lendar and its users as the only acceptable way to

change people's behavior and attitudes.

12

Conclusion

The Shower Calendar offers a way to foster awareness

of water consumption and communication among family

members. This ultimately has the potential to result in

behavioral change. Specifically, participants mentioned

the Calendar's appeal to their ambition to change,

which resulted in personal goal setting. Beyond result-

ing feelings of achievement, competence and control

due to the improvement, its "social features" (indivi-

dualization, persistence) led to communication and

competition, and in the case of success, to feelings of

popularity and acceptance.

While many of the intended experiential effects became

apparent, actual behavioral change was less consistent.

Family 1 reduced their water consumption significantly

over the 31 days, while in family 2 only the mother

showed a marginally significant reduction, the sons did

not respond. This highlights a common problem with

persuasive technologies. We understand them as "ma-

terialized arguments" [17] – dialog partners and media-

tors rather than educators or watchdogs. Thus, their

general design assumes that people have an "inner

wish" to change, that is, to transform themselves. In-

deed, Fogg [7, p. 15) defines persuasion as the "at-

tempt to change attitudes or behaviors or both (without

using coercion or deception)". This emphasizes volunta-

riness and a certain wish to transform, but it under-

plays the active participation of the one to be per-

suaded. The Shower Calendar doesn't coerce those who

don't want to change. While it embodies a clear objec-

tive – to reduce water consumption – it does not conti-

nually call the user's behavior into question. We believe

this to be an important attitude in the design of persua-

sive technologies. Regulation and automation may

promise more success in terms of water saved, but

according "products" might be experienced as patroniz-

ing, judgmental, or inhumane. Instead of providing a

meaningful, stimulating and positive context for con-

servation, it may become rather loaded with negative

experiences and emotions. And if coerced, people will

definitely find a different outlet for their need to in-

dulge.

The problem that arises from this view, however, is the

question of how we should evaluate persuasive tech-

nologies? We cannot claim to create products that fos-

ter change if we cannot actually demonstrate change.

This carries the danger of arbitrariness, a "whatever-

ism", where every idea is good, even if this goodness

cannot be demonstrated. The present case attempts to

avoid this. It understands the product as a theory (see

[3], [21], [14]), which has to be developed out of the

discussion of previous theories (i.e., products), which

has to be plausible in itself, and which makes expecta-

tions explicit. In addition, we must not only focus on

change itself, but also on the experiences accompany-

ing this change. The objective is not only to demon-

strate and maximize change in behavior and attitude,

but to reveal boundary conditions, and to make change

a worthwhile experience.

In the present case, we could demonstrate that the

Shower Calendar did not coerce but fostered communi-

cation and competition in both families. However, Fami-

ly 1 was more successful in translating this offer into

behavioral change. This reminds us of the fact that

change is not achieved by the product itself (as in au-

tomation or regulation) but by the people involved. In

this sense, we prefer talking about Transformational

Products rather than persuasive technologies. Trans-

formational products offer ways for self-improvement

13

and support people in their wish to transform (see [20],

[13], [15]), but do so by emphasizing the active role of

the individual in the process. This active role is crucial.

There is simply a difference between changing the be-

havior of somebody and enabling somebody to change

herself.

The experiences participants reported can be traced to

the Shower Calendar's specific characteristics and fea-

tures. No immediate reduction of water-consumption

without its direct feedback; no goal setting, compari-

son, communication, and competition without persis-

tent and individualized feedback. No positive feelings

without the careful design of a non-judgmental, non-

coercive feedback. This highlights the intimate relation-

ship between the detailed, formal design and resulting

effects. While broad principles, such as goal setting and

feedback, are important guides, their translation into a

material form is crucial. Case studies, such as the

present and those cited, enable the specific critical

discourse and the concrete empirical exploration of

abstract principles and their materialization. This makes

them worthwhile.

Acknowledgements

This work was supported by "Universal Home", a

project of Gira, Miele, Poggenpohl, RWE, Schott, Vail-

lant, WMF, and 3M in cooperation with the Zollverein

Foundation.

References [1] Arroyo, E. and Bonanni, T. S. WaterBot: Exploring Feedback and Persuasive Techniques at the Sink. In Proc. CHI 2005, ACM Press (2005), 631-639.

[2] Brehm, J.W. Theory of psychological reactance.

Academic Press, New York, USA, 1966.

[3] Carroll, J. M., Singley, M. K., and Rosson, M. B. Integrating theory development with design evaluation. Behaviour & Information Technology, 11 (1992), 247–255.

[4] Cialdini, R. The science of persuasion. Scientific American (2001), 76-81.

[5] Darby, S. The effectiveness of feedback on energy consumption. University of Oxford. http://www.eci.ox.ac.uk/research/energy/downloads/smart-metering-report.pdf [online], 2006.

[6] Fishbein, M. and Ajzen, I. Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research. Addison-Wesley, Massachusetts, 1975.

[7] Fogg, B. J. Persuasive Technology: Using Comput-ers to Change What We Think and Do. Morgan Kauf-mann, San Francisco, CA, USA, 2002.

[8] Froehlich, J., Findlater, L. and Landay, J. The De-sign of Eco-Feedback Technology. In Proc. CHI 2010, ACM Press (2010), 1999-2008.

[9] Gustafsson, A. and Gyllenswärd, M. The power-

aware cord: energy awareness through ambient infor-mation display. Ext. Abstracts CHI 2005, ACM Press (2005), 1423-1426.

[10] He, HA., Greenberg, S. and Huang, EM. One Size Does Not Fit All: Applying the Transtheoretical Model to Energy Feedback Technology Design. In Proc. CHI 2010, ACM Press (2010), 327-936.

[11] Kappel, K. and Grechenig, T. "show-me": water consumption at a glance to promote water conservation in the shower. In Proc. PERSUASIVE 2009, ACM Press (2009), Article No. 26.

[12] Kuznetsov, S. and Paulos, E. UpStream: motivating

water conservation with low-cost water flow sensing and persuasive displays. In Proc. CHI 2010, ACM Press (2010), 1851-1860.

[13] Laschke, M., Diefenbach, S., Heidecker, S. and Hassenzahl, M. Transformationale Produkte: Acht Kon-

14

zepte zum schonenden Umgang mit Ressourcen. In Proc. Mensch und Computer 2010 (2010), 189-194.

[14] Ozenc, F. K., Transitions Research for Experience Design. International Association of Societies of Design

Research (IASDR 2009), 2009, Seoul, Korea.

[15] Pine II, B. J. and Gilmore, J. H. The Experience Economy. Harvard Business School Press, Boston, MA, USA, 2009.

[16] Rachlin, H. The Science of Self-Control, Harvard University Press, Cambridge, Massachusetts, USA, 2004.

[17] Redström, J. Tangled interaction: On the expres-siveness of tangible user interfaces. ACM Transactions on Computer-Human Interaction, 15, 4 (2008), Article No.16.

[18] Umweltbundesamt. Umweltdaten Deutschland – Nachhaltig wirtschaften – Natürliche Ressourcen und Umwelt schonen. Umweltbundesamt, Dessau, Germa-ny, 2007.

[19] Wikipedia, „Wasserverbrauch“ (water consumption) http://de.wikipedia.org/wiki/Wasserverbrauch [online], 2010.

[20] Zimmerman, J. Designing for the Self: Making products that help people become the person they de-sire to Be. In Proc. CHI 2009, ACM Press (2009), 395-404.

[21] Zimmerman, J. Forlizzi, J. and Evenson, S. Re-search through design as a method for interaction de-sign research in HCI. In Proc. CHI 2007, ACM Press (2007), 493-502.